This directory contains scripts and test data for experimenting with PIDC network inference, introduced in Gene Regulatory Network Inference from Single-Cell Data Using Multivariate Information Measures (Cell Systems, 2017).
Data were simulated using GeneNetWeaver, as described in Methods. Our analyses on experimental datasets can be reproduced by downloading the datasets from the relevant references, putting them into the same format as our simulated datasets, and pointing the Infer network script to them.
goldstandards: True network structures, from GeneNetWeaver, to compare with inferred networks.
networks: Networks inferred from the various datasets, using the various algorithms. (NB The R package MINET can be used to reproduce our results for the ARACNE, CLR and MINET algorithms.)
notebooks: Scripts for inferring networks and speed benchmarking.
simulated_datasets: Single-cell datasets, simulated using GeneNetWeaver, from the goldstandards.
Julia 0.6
InformationMeasures (version 0.1.1 or higher)
Pkg.add("InformationMeasures")
PyPlot
Pkg.add("PyPlot")
LightGraphs
Pkg.add("LightGraphs")
GraphPlot
Pkg.add("GraphPlot")
NetworkInference
Pkg.add("NetworkInference")